tracted by two projecting points. Three miles above the one on the North side, and abreast of two small isles, is the passage out to sea, or to the North entrance, and this lies nearly in the direction of N. by W. and S. by E.

The North entrance lies in lat. 45° 38′ S., and 5 leagues to the North of Five Fingers Point. To make this entrance plain, it will be necessary to approach the shore within a few miles, as all the land within and on each side is of considerable height. Its situation may, however, be known at a great distance, as it lies under the first craggy mountains which rise to the North of the land of Five Fingers Point. The southernmost of these mountains is remarkable, having at its summit two small hillocks. When this mountain bears S.S.E. you will be before the entrance, on the South side of which are several isles. The westernmost and outermost is the most considerable, both for length and circuit; and this I have called Break Sea Isle, because it effectually covers this entrance from the violence of the S.W. swell, which the other entrance is so much exposed to. In sailing in you leave this isle, as well as the others, to the South. The best anchorage is in the first or North arm, which is on the larboard hand going in, either in one of the coves, or behind the isles that lie under the S.E. shore.

A

The country is exceedingly mountainous, not only about Dusky Bay, but through all the southern part of this western coast of Tavai Poenammoo. prospect more rude and craggy is rarely to be met with; for inland appears nothing but the summits of mountains of a stupendous height, and consisting of rocks that are totally barren and naked, except where they are covered with snow. But the land bordering on the sea-coast, and all the islands, are thickly clothed with wood, almost down to the water's edge. The trees are of various kinds, and many of them are from 6 to 8 and 10 feet in girth, and from 60 to 80 or 100 feet in height.

ANCHOR ISLAND HARBOUR, as its name indicates, lies on the island in the mouth of Dusky Bay, on the North side of the island. It was in this harbour that Vancouver's ship, the Discovery, rode out safely a tremendous gale from N.W. by W., November 3-6, 1791. He states:-It appeared to be perfectly secure, and may be found convenient when accident may prevent vessels from getting into Facile Harbour. It has two entrances; that to the North of the Petrel Islands is a fair and clear channel, though of great depth, from 33 to 38 fathoms. In the narrowest part it is about a cable's length wide, and, I believe, free from any danger, as the shores are steep, without any sunken rocks or shoals, excepting within the passage close under the South side of Large Petrel Island, shown by weeds, and out of the way. The other passage is to the southward of the Petrel Islands, and as in all probability a strong northerly wind would alone induce any person to make choice of this in preference to Facile Harbour, the S.W. point of Large Petrel Island should be kept close on board, which may be safely done, in order to weather the rock that appears above water in the middle of the harbour, and to avoid a sunken one, of which there is not the least indication, and on which there is no greater depth than 12 feet at low water. Between this sunken rock and the point off which it lies, about three-quarters of a cable's length, and nearly in the direction to what I have called Entry Island, are 16 fathoms. Keeping the rock in the harbour, which is always visible, in a

line with what I have called North Entry Island, will be a sufficient direction to pass within the above-mentioned point and the sunken rock. Capt. Cook's chart of the port is excellent, and the only farther remark necessary is, that Anchor Island Harbour, although a very safe and secure port, is not a very convenient one to get to sea from, owing to its narrow limits, great depth of water, and the above sunken rock on its western entrance.

The WEST COAST of the Middle Island, between these last described bays and Cape Farewell, is almost a terra incognita in a nautical view. We have but the very vague accounts of passing observers in the South, or the remarks made by travellers on land for the northern section, neither of which can be of any service to the seaman. But very little can be said here on it.

For the space of nearly 250 miles from Dusky Bay the charts contain almost the entire information we possess. It is said there are some good harbours in the interval; but the rivers must be of small extent, because the mountain ranges here approach the shore. Doubtful Harbour (Point February, or South Point, about lat. 45° 12', lon. 167° 0′), Mary's Bay, Milford Haven, lat. 44° 32', lon. 167° 45', and Cascade Point, lat. 43° 55', lon. 168° 30′, are the most prominent features.

BOLD HEAD, lat. 42° 57', lon. 170° 40′ E., was reached by a land expedition under Messrs. Brunner and Heaphy in March, 1846. They started from Nelson, and to their remarks the delineation of the coast is owing. For about 30 miles from Bold Head the coast is low, with a sandy beach. In this interval are the mouths of three rivers, the Arahura, or Greenstone, or Brunner, the Teramakau, and the Gray River, or Mawhera; the last is the principal. Its mouth is in about lat. 42° 39′, and is unavailable for any but small craft. The mountains between the valley of the Gray and that of the Greenstone do not approach within 30 miles of the coast; and a few natives, who live at the mouth of the latter river, report that there is an easy communication by the valley which it forms across the island to Banks's Peninsula.

At about 35 miles North of the Gray is a group of peaks, called by Cook, and also by D'Urville, the Five Fingers. The coast in this space is iron-bound, with the exception of a beach about 8 miles long, not backed by any level country.

CAPE FOULWIND, a projecting part of the coast, lies about 16 miles North of the Five Fingers. The land about it is high and rocky. Off it are the rocks called the Three Steeples. The cape is low and covered with trees, lat. 41° 46′ S., lon. 171° 29′ E. The coast from Cape Foulwind trends first to the West, then N.E., and then due North; it is low and sandy, and in the bight of the bay thus formed is the mouth of the Buller River or Kawatiri, about 10 miles West from Cape Foulwind. Although larger than any of the rivers falling into the sea between this and Cape Farewell, the Buller appears to be unavailable for any but small craft, having a shallow entrance.

At 18 miles N.E. from the Buller River is the mouth of the Ngakuhau, North of which is a sandy beach, to which, however, the mountains approach closely. At this distance is the mouth of the Mokihinui River. The coast has an iron

* Voyage de L'Astrolabe, vol. ii. p. 345.

bound character for about 10 miles in a N.N.E. direction to a sandy beach, 16 miles in extent, through which the River Karamea flows.

ROCKY POINT is in lat. 40° 54', lon. 172° 10′ E. The coast for about 14 miles to the South of it trends nearly due North and South, a snowy range approaching close to the sea. Numerous small streams have their mouths along this coast, and there are two larger ones, the Wahapoai or Heaphy, 5 miles South of Rocky Point, and the Haihai, at its southernmost extremity. From Rocky Point to the N.E., for a distance of 35 miles, to the South Wanganui, the coast is rocky and iron-bound, steep spurs from the high mountains, composed of granite and gneiss, come down into the sea.

SOUTH WANGANUI is a very good harbour for small vessels, about 10 miles S.W. of Cape Farewell. Here there are extensive coal beds, equally easy of access with those in Massacre Bay, which are probably portions of the same formation. CAPE FAREWELL, the northernmost point of the Middle Island, has been before described on page 760, and this completes the circuit of the group.*

It will be seen that many of the details given of this important country are exceedingly vague, far more so than its growing importance would demand; but as the requirements of the mariner will increase with the progress of colonization, all farther information must be acquired progressively.

CHAPTER XXII.

ISLANDS BETWEEN LATITUDES 20° AND 40° SOUTH.

THE immense extent of ocean comprised in this belt of the South Pacific, between the coasts of Chile and Australia, has but few, very few, spots of land in the southern portion. In the northern half the coral groups of the Low Archipelago extend to the South of, or within, our present limits, but will be described in a separate section hereafter. In pursuance of the principle stated on page 652, we return from the description of New Zealand in the West to the lower latitude on the South American coast, and thence again proceed to the westward.

JUAN FERNANDEZ ISLANDS.

This group, if it can be so designated, consists of two chief islands, at a considerable distance asunder, with some smaller ones attached to each. Their name is derived from Juan Fernandez, a Spaniard, who discovered them in his voyage from Lima to Valdivia, in 1563. He designed to settle here, and requested a patent for them, but did not obtain it. It was much visited by the buccaneers in

Some rocks, called the Farewell Rocks, have been marked on some charts as lying about 10 miles to the northward of Cape Farewell, but it is stated that on a good lookout they could not be found. See Nautical Magazine, 1842, p. 810.

and

their marauding expeditions against the Spaniards, and Dampier describes his visits here. In 1681 a Mosquito Indian was left on it by the ship Dampier was in, and was taken off by him in March, 1684, he having lived three years solitarily, upon the goats first introduced by Fernandez. In February, 1709, Capt. Woodes Rogers touched here, and found the well-known hero, Alexander Selkirk, who had been left on it by the ship Cinque Ports, Capt. Stradling, four years four months previously, Dampier having been both in the ship he was landed from and that which took him off. Selkirk gave an account of his sojourn to Daniel Defoe, in order to prepare them for the press, and from the ideas there given the excellent romance of Robinson Crusoe was formed. In the visits above mentioned, a prodigious number of goats were found, but these were reduced by the Spaniards, who introduced dogs for the purpose of destroying one of the principal refreshments of their enemies the buccaneers. These animals were both found in possession of the island by Lord Anson, with the miserable remnant of his crews in the Centurion, the Tryal sloop, and the Gloucester, in June, 1741. Every one is familiar with the terrible descriptions of the havoc made by the scurvy in these ships, and the restoration of the survivors to health at Juan Fernandez. Men thus saved from utter destruction would naturally paint in glowing colours the scene of their deliverance; and these accounts leading to the supposition that the English would colonize both Inchin, in the Chonos Archipelago, and Juan Fernandez, the Spaniards sent a colony in 1751, but it was soon after almost totally destroyed by the dreadful earthquake, a calamity the island has since been subject to on more than one occasion. It was still inhabited when Carteret visited it in 1769. In 1819 the Chilian government formed it into a penal colony, but it was not much kept up, on account of the expense. When Capt. P. P. King, in H.M.S. Adventure, was here in 1830, it was occupied, or rather rented from the Chilian government for a term of years. In 1835 this prison colony had been increased, and the prisoners rose on, and for a short time overcame, the troops. This fact was related by Capt. P. Masters, which led to a counter-statement from the ex-governor, Mr. T. Sutcliffe, an Englishman. After this it was deserted, and at this period there is only one family, who supply the whalers and other visitors with firewood and goats' flesh, and also serve as guides to the island. The following notices have been gathered from the works quoted below.

The principal islands are Juan Fernandez, called for distinction Mas-a-Tierra, because it is nearest the continent, and Mas-a-Fuera, or "more in the distance," which is about 90 miles West of it.

JUAN FERNANDEZ is about 10 or 12 miles long by 6 miles broad. It has

Dampier's Collection, vol. i. p. 83, et seq.

+ The numerous accounts of this island may be perused with much interest. Besides the extracts given in Burney's and other collections, they may be enumerated thus :-Dampier's Collection, vol. i. pp. 84-92, a good account; Woodes Rogers' Account of the Island and Alexander Selkirk; in Lord Anson's Voyage Round the World; Noticias Secretas de America, por Dons J. G. and A. de Ulloa, pp. 50-56, a good account; Bennett's Whaling Voyage; Voyage of the Adventure and Beagle, vol. i. pp. 302-308; Geographical Journal, vol. iv. p. 183; Nautical Magazine, June, 1837, pp. 360-362; The Earthquake of Juan Fernandez in 1835, by the Retired Governor (T. Sutcliffe), Manchester, 1839; Four Years in the Pacific, by the Hon. Fred. Walpole, 1849, vol. i. chap. 14, p. 353, et seq.; and numerous other authors.

several bays, and its general appearance is thus described by Capt. P. P. King, who anchored in Cumberland Bay, on its North side, January, 1830.

"I have seldom seen a more remarkable and picturesque view than is presented by the approach to Juan Fernandez. When seen from a distance, the mountain of the Yungue' (anvil), so called from its resemblance to a blacksmith's anvil, appears conspicuously placed in the midst of a range of precipitous mountains, and is alone an object of interest. It rises 3,000 feet above a shore, which is formed by an abrupt wall of dark-coloured bare rock, 800 or 900 feet in height, through whose wild ravines, broken by the mountain torrents, views are caught of verdant glades, surrounded by luxuriant woodland.

"The higher parts of the island are, in general, thickly wooded, but in some places there are grassy plains of considerable extent, whose lively colour contrasts agreeably with the dark foliage of myrtle trees which abound on the island.

"The Yungue is wooded nearly from the summit to its base, whence an extensive and fertile valley'extends to the shore, and is watered by two streams, which take their rise in the heights and fall into the sea.

"This valley appears to have been formerly cleared and cultivated by the Spaniards, who had a colony here; for the stone walls, which served to divide their enclosures, still remain. From Walter's Account of Anson's Voyage, and the view given with it of the commodore's tent, there is no difficulty in determining this valley to be the spot on which his encampment was placed.

"The dwellings of the settlers are erected on the flat land at the North side of the bay, where the soil is richer than in other parts, and where it is more sheltered from the squalls which, during strong southerly gales, rush down the valley of the Yungue, the situation of the former establishment, with great violence.

"The remains of a fort, called San Juan Bautista, are yet in a tolerable state; it is situated on a rising ground, about 130 feet above the sea, at the S. W. part of the bay, and overlooks the village. In the middle of the beach are some ruins of a four-gun battery, and there are traces of a fort at the N.W. end of the island. "By sending a boat to the East point of the bay, to fish in 40 fathoms water, a most delicious kind of cod-fish may be taken in such numbers, that two men in half an hour could fill the boat. Crawfish, of large size, are almost equally abundant; they are taken with a hooked stick; one of our boats caught forty-five in a very short time. The inhabitants catch them and cure their tails, by exposure to the sun, for exportation to Chile, where they are much esteemed, and fetch a high price.

"Fort Juan Bautista is in lat. 33° 37' 45" S., lon. 78° 53′ W."

The next extracts are from the notes of Capt. Peter Masters, of Liverpool, who came hither in 1835.

French Bay is small, and the curve which forms it differs but little from the direction of the outer part of the coast. It affords no shelter, and is situated at the bottom of a valley which rises quickly to the high land, presenting a pleasing appearance, being well covered with trees, some of which were very large. Between French and Cumberland Bays there appeared to be no landing place.

Cumberland Bay lies exposed to northerly winds, and in the months when these winds prevail no ship should anchor in it; it curves in from the heads which form